Image-to-Text
PyTorch
Safetensors
PEFT
English
remote-sensing
satellite-imagery
earth-observation
change-detection
visual-grounding
image-captioning
visual-question-answering
optical-sar-fusion
sar
multimodal
lora
Instructions to use thundercode/SatQuery with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use thundercode/SatQuery with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
release: add docs/TESTING.md
Browse files- docs/TESTING.md +1 -1
docs/TESTING.md
CHANGED
|
@@ -1172,7 +1172,7 @@ harness outputs (`run_output.txt`, `run_final2.txt`, `run_final3.txt`), the reco
|
|
| 1172 |
**24 live runs, 24 correct dispatches, 0 mock nodes, no run id repeated across passes.** The trace fill
|
| 1173 |
was **94.4444 %** on every case, and every case carries a real `run_*` id, the Hugging Face link in the
|
| 1174 |
DOM, and the `capabilities → assets → infer` call sequence all addressed to
|
| 1175 |
-
`
|
| 1176 |
|
| 1177 |
The eight cases are Phase A (six regression cases: `vqa`, `caption`, `grounding`, `change`,
|
| 1178 |
`change_vqa`, `optical_sar`) and Phase B (the two defect cases: `"Where are the built-up areas in this
|
|
|
|
| 1172 |
**24 live runs, 24 correct dispatches, 0 mock nodes, no run id repeated across passes.** The trace fill
|
| 1173 |
was **94.4444 %** on every case, and every case carries a real `run_*` id, the Hugging Face link in the
|
| 1174 |
DOM, and the `capabilities → assets → infer` call sequence all addressed to
|
| 1175 |
+
`<backend-host>` (two `assets` calls for the pair tasks).
|
| 1176 |
|
| 1177 |
The eight cases are Phase A (six regression cases: `vqa`, `caption`, `grounding`, `change`,
|
| 1178 |
`change_vqa`, `optical_sar`) and Phase B (the two defect cases: `"Where are the built-up areas in this
|